Our combined in vitro biochemical and structural studies on a representative member of the IS200/IS605 transposase family demonstrated that this family uses a completely novel recombination pathway involving the movement of only single-stranded DNA. One particularly surprising discovery was that the transposase recognizes its target site through DNA-DNA interactions rather than using a site-specific DNA binding domain: target site recognition is accomplished by base pairing interactions between the target site and an internal segment of transposon DNA. This suggests the possibility that by changing the internal segment, targeting could be directed to novel target sites. If we can do this, this might allow the precise introduction of exogenous genes into benign locations in chromosomes or places where gene expression can be appropriately controlled in a cell- and development-specific manner. In our recent work, we have been continuing to explore the mechanism of IS200/IS605 transposition. In particular, we have been investigating how the number of nucleotides between the transposon ends and the recognition DNA hairpin (the linker length) affects IS608 transposition, and also how a proposed structural change drives the process from DNA strand cleavage to strand transfer. Our data is consistent with our previously proposed rotation model in which two flexible alpha-helices alternate their configuration with respect to the enzyme active sites, and that the back-and-forth between these configurations - along with a reset step - drives the transposition reaction forward. We have also been studying the putative transposase associated with bacterial Repeated Extragenic Palindromic Sequences (or REPs). REPs form nucleotide stem-loop structures and are found scattered in high numbers in many bacterial species. Their sheer number suggests there was a process that led to their expansion in their host species, and it has been proposed that this might involve an protein closely related to the IS200/IS605 transposases. To confirm this, we determined the structure of the TnpA(REP) from E. coli strain MG1655 in complex with a DNA palindrome. Indeed, it resembles the IS200/IS605 transposases and shares the property of being able to cleave certain DNA structures that contain REP sequences. Thus, it appears likely that it has been responsible for the proliferation of REP sequences throughout bacterial genomes, and has been an important contributor to genome evolution. Curcio, M.J. and Derbyshire, K.M. (2003) Nat. Rev. Mol. Cell. Biol. 4, 865-877. Debets-Ossenkopp, Y.J., et al. (1999) Antimicrob. Agents Chemother. 43, 2657-2662. Kersulyte, D., et al. (2002) J. Bacteriol. 184, 992-1002. Mennecier, S., Servant, P., Coste, G., Bailone, A., and Sommer, S. (2006) Mol. Microbiol. 59, 317-325. Sebaihia, M. et al. (2006) Nature Genet. 38, 779-786.